Talk @ Europython, Rimini, Italy 13/07/2017 - AI: The Story So Far - Amazon AI: Lex, Polly, Rekognition - Apache MXNet overview - Apache MXNet demos - Tools and Resources

1. © 2015, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Julien Simon, Principal Technical Evangelist, AWS
@julsimon
AI on a Raspberry Pi

2. Agenda
• AI: The Story So Far
• Amazon AI
• Apache MXNet overview
• Apache MXNet demos
• Tools and Resources

4. • Machine Learning is now a commodity, but still no HAL in sight
• Traditional Machine Learning doesn’t work well with problems
where features can’t be explicitly defined
• So what about solving tasks that are easy for people to
perform, but hard to describe formally?
• Is there a way to get informal knowledge into a computer?
Where is HAL?

5. • Universal approximation machine
• Through training, a neural network
discovers features automatically
• Not new technology!
• Perceptron - Rosenblatt, 1958
image recognition, 20x20 pixels
• Backpropagation - Werbos, 1975
• They failed back then because:
• Data sets were too small
• Solving large problems with fully connected networks
required too much memory and computing power,
aka the Curse of Dimensionality
Neural Networks, Revisited

6. Everything is digital: large data sets are available
• Imagenet: 14M+ labeled images - http://www.image-net.org/
• YouTube-8M: 7M+ labeled videos - https://research.google.com/youtube8m/
• AWS public data sets - https://aws.amazon.com/public-datasets/
The parallel computing power of GPUs make training possible
• Simard (2005), Ciresan (2011)
• State of the art networks have hundreds of layers
• Baidu’s Chinese speech recognition: 4TB of training data, +/- 10 Exaflops
Cloud scalability and elasticity make training affordable
• Grab a lot of resources for fast training, then release them
• Using a DL model is lightweight: you can do it on a Raspberry Pi
Why It’s Different This Time

7. ImageNet Large Scale Visual Recognition Challenge (ILSVRC)
Same breed?
Humans: 5,1%

8. Amazon AI

9. Fulfilment &
Logistics
Search &
Discovery
Existing
Products
New
Products
Thousands of Amazon Engineers Focused on AI

10. Amazon Echo

11. Infrastructure CPU
Engines MXNet TensorFlow Caffe Theano Pytorch CNTK
Services
Amazon Polly
Chat
Platforms
IoT
Speech
Amazon Lex
Mobile
Amazon AI: Artificial Intelligence In The Hands Of Every Developer
Amazon
ML
Spark &
EMR
Kinesis Batch ECS
GPU
Amazon Rekognition
Vision

12. Amazon Polly: Text To Speech
Powered By Deep Learning
Amazon Polly
“The temperature
in WA is 75°F”
“The temperature
in Washington is 75 degrees
Fahrenheit”
Text In, Life-like Speech Out

13. Amazon Lex
Speech Recognition & Natural Language Understanding
Amazon Lex
Automatic Speech Recognition
Natural Language Understanding
“What’s the weather
forecast?”
Weather
Forecast
“It will be
sunny
and 75F”
“It will be sunny
and 75 degrees
Fahrenheit”
Amazon Polly

14. Amazon Rekognition
Image Recognition And Analysis Powered By Deep Learning
Amazon Rekognition
Car
Outside
Daytime
Driving
Objects/Scenes
Female
Smiling
Sunglasses
Faces
Images In, Categories and Facial Analysis Out

15. Demo #1 – Amazon Polly & Rekognition
https://medium.com/@julsimon/a-hands-on-look-at-the-amazon-rekognition-api-e30e19e7d88b
https://medium.com/@julsimon/amazon-polly-hello-world-literally-812de2c620f4
https://github.com/juliensimon/aws/tree/master/rekognition

16. Artificial Intelligence on AWS today

17. Apache MXNet Overview

18. Apache MXNet
Programmable Portable High Performance
Near linear scaling
across hundreds of GPUs
Highly efficient
models for mobile
and IoT
Simple syntax,
multiple languages
Most Open Best On AWS
Optimized for
deep learning on
AWS
Accepted into the
Apache Incubator

19. 0
4
8
12
16
1 2 4 8 16
Ideal
Inception v3
Resnet
Alexnet
91%
Efficiency
Multi-GPU Scaling With MXNet

20. Ideal
Inception v3
Resnet
Alexnet
88%
Efficiency
0
64
128
192
256
1 2 4 8 16 32 64 128 256
Multi-Machine Scaling With MXNet

21. Apache MXNet demos

22. Demo #2 – Training MXNet on MNIST
https://medium.com/@julsimon/training-mxnet-part-1-mnist-6f0dc4210c62
https://github.com/juliensimon/aws/tree/master/mxnet/mnist

23. Demo #3 – Object Detection on a Raspberry Pi
https://medium.com/@julsimon/an-introduction-to-the-mxnet-api-part-6-fcdd7521ae87
GoPiGo Arduino Yùn AWS IoT
MQTT@CallMeJohnnyPi

24. One-Click GPU or CPU
Deep Learning
AWS Deep Learning AMI
Up to~40k CUDA cores
Apache MXNet
TensorFlow
Theano
Caffe
Torch
Keras
Pre-configured CUDA drivers,
MKL
Anaconda, Python3
Ubuntu and Amazon Linux
+ CloudFormation template
+ Container Image

25. Additional Resources
MXNet Resources
• MXNet Blog Post | AWS Endorsement
• Read up on MXNet and Learn More: mxnet.io
• MXNet Github Repo
• MXNet Recommender Systems Talk | Leo Dirac
AWS Resources
• Deep Learning AMI |Amazon Linux
• Deep Learning AMI | Ubuntu
• CloudFormation Template Instructions
• Deep Learning Benchmark
• MXNet on Lambda
• MXNet on ECS/Docker https://soundcloud.com/amazon-web-services-306355661/
200-introduction-to-apache-mxnet-on-aws

26. Grazie mille!
Julien Simon, Principal Technical Evangelist, AWS
@julsimon